Logistics

• Hours: 10-11:15AM
• Room: G.K.1.4-Guarini Campus, Kushlan Wing, First Floor, Room 4
• Office House: By appointment

Learning Outcomes

Upon successful completion of this course the students will be able to: ::: fragment * execute basic programming tasks in R (e.g. loops, conditional statements, while statements, etc.) ::: ::: fragment * understand basic GIS terms and concepts ::: ::: fragment * utilize GIS for conducting spatial analyses. ::: ::: fragment * appreciate the design and structure of a geographic information system (GIS) as a decision-making tool. ::: ::: fragment * produce maps :::

Grading

You will be graded on four problem sets during the semester (each 12.5% of your grade) and a final report and presentation (each 25% of your grade). ::: fragment * 4 problem sets: 50% of the final grade (12.5% each) * final presentation: 25% of the final grade * final project report: 25% of the final grade :::

Problem Sets

Final Project

Books to Use: Data Analysis and Visualization

Books to Use: GIS

Other useful Sources

• A Guide on Data Analysis

• Intro to GIS and Spatial Analysis

• Introduction to Spatial Data Programming with R

Why R

R is good for:

• Automation - doing unfeasible repetitive tasks
• Reproducibility - using the same commands repeatedly and obtaining the same output
• Visualization - making and presenting graphs and maps

Processing Vector Layers sf

sf will be the main library that we will work with

It will help us deal with:

• Numerical Operations to calculate: Areas, Length, Distances, etc.
• GIS Logical Operations: Contains, Within, Within distance, Crosses, Overlaps, Equals, Intersects, etc.
• Geometry Operations: Centroid, Buffer, Intersection, Union, Difference, Convex-Hull, Simplification, etc.

Processing Vector Layers sf: buffer

Processing Raster Layers stars

We can perform geometric operation on rasters (pictures) with the stars package

• Accessing cell values - as a matrix or as a dataframe, extracting pixels to points
• Performing Raster algebra: raster arirthmentic and logic
• Changing the resolution and extent: cropping, mosaicing, resampling, and reprojecting
• Transforming Rasters: to points and polygons

Processing Raster Layers stars

Temperature in 1901

Processing Raster Layers stars

Temperature in 2022

Processing Raster Layers stars

Temperature difference between 2022 and 1901 > 4

Geometric calculations on longitude/latitude with geosphere

Add more projects here

Data visualization

• ggplot2 is the library that will allow to visualize data analysis results, but also to make maps
• it has well designed and consistent syntax that supports visualization for both vectors and rasters
• it has highly customizable publication-quality figures and maps

Interactive Data visualization

• leaflet is a library that allows us to make interactive maps
• mapview is a wrapper around leaflet automating the addition of: labels, popups, color scales, and common basemaps
• see the restaurants in central Rome below

Programming

• A programming language is a machine-readable artificial language designed to express computations that can be performed by a computer.
• Programming allows us to edit code and re-use it in the future and obtain the same results in the future

Object-oriented programming

• In object-oriented programming, the interaction with the computer takes place though objects
• Each object belongs to a class: an abstract structure that has specific properties

Example:

• All cars in the parking lot are instances of the “car” class
• The “car” class has specific properties: make, color, year and methods: start, drive stop

Object-oriented programming

Object-oriented programming

Object-oriented programming

Object-oriented programming in R

We will see that everything that we work with in R is an object

For example, we can load up a geojson file in R.

#Step1: Loading the geojson file
restaurants <- geojson_sf("/Users/bgpopescu/Library/CloudStorage/Dropbox/john_cabot/teaching/big_data/week13/data/restaurant.geojson")
#Step2: Selecting only the relevant variables
restaurants<-subset(restaurants, select = c(name, `addr:street`))
#Step3: Removing the restaurants without a name or without an address
restaurants2<-subset(restaurants, !is.na(restaurants$name) | !is.na(restaurants$`addr:street`))

R transforms the geojson file into an object of a class named sf data.frame

#Step4: Identifying the object class
class(restaurants2)

[1] "sf"         "data.frame"

Object-oriented programming in R

This type of object has numerous properties such as:

• rows

nrow(restaurants2)

[1] 2811

• columns

ncol(restaurants2)

[1] 3

• Coordinate Reference System (CRS)

st_crs(restaurants2)

Coordinate Reference System:
  User input: 4326 
  wkt:
GEOGCS["WGS 84",
      DATUM["WGS_1984",
        SPHEROID["WGS 84",6378137,298.257223563,
          AUTHORITY["EPSG","7030"]],
        AUTHORITY["EPSG","6326"]],
      PRIMEM["Greenwich",0,
        AUTHORITY["EPSG","8901"]],
      UNIT["degree",0.0174532925199433,
        AUTHORITY["EPSG","9122"]],
      AXIS["Latitude",NORTH],
      AXIS["Longitude",EAST],
    AUTHORITY["EPSG","4326"]]

Object-oriented programming in R

Once imported, the sf data.frame is saved in the computer memory

Printing the object will display some of its properties and specific properties

restaurants2

Simple feature collection with 2811 features and 2 fields
Geometry type: POINT
Dimension:     XY
Bounding box:  xmin: 12.21167 ymin: 41.70574 xmax: 12.77428 ymax: 42.06974
Geodetic CRS:  WGS 84
First 10 features:
                            name                 addr:street
1              Pizzeria ai Marmi         Viale di Trastevere
2                  Sichuan Haozi Via di San Martino ai Monti
3  Dar filettaro a Santa Barbara           Largo dei Librari
4                 Al Peperoncino                Via Ostiense
6                 Ai Tre Scalini              Via Panisperna
7          Trattoria Ada e Mario      Circonvallazione Appia
8                     Gustosando                        <NA>
9                      Sa Posada            Via Elvia Recina
10            Pizzeria Formula 1              Via degli Equi
11                  Da Francesco             Piazza del Fico
                    geometry
1  POINT (12.47379 41.88826)
2   POINT (12.49948 41.8958)
3   POINT (12.4737 41.89467)
4  POINT (12.47698 41.85343)
6  POINT (12.49044 41.89628)
7  POINT (12.51433 41.87532)
8  POINT (12.42743 41.89954)
9   POINT (12.5079 41.87995)
10 POINT (12.51268 41.89702)
11  POINT (12.4704 41.89932)

Object-oriented programming in R

By printing the object, we can see some of its properties including:

• dimension

• bounding box

• crs

Inheritance

One of the characteristics of object oriented programming is inheritance

Inheritance is what makes it possible for one class to extend to another class, by adding other properties

Example:

• A “taxi” is an extension of a “car” class, inheriting all of its properties and methods.

• A taxi could have new properties like taxi company name

Inheritance

In R, every complex object is a collection of smaller components such a properties

We can use str to examine the properties of the class

str(restaurants2)

Classes 'sf' and 'data.frame':  2811 obs. of  3 variables:
 $ name       : chr  "Pizzeria ai Marmi" "Sichuan Haozi" "Dar filettaro a Santa Barbara" "Al Peperoncino" ...
 $ addr:street: chr  "Viale di Trastevere" "Via di San Martino ai Monti" "Largo dei Librari" "Via Ostiense" ...
 $ geometry   :sfc_POINT of length 2811; first list element:  'XY' num  12.5 41.9
 - attr(*, "sf_column")= chr "geometry"
 - attr(*, "agr")= Factor w/ 3 levels "constant","aggregate",..: NA NA
  ..- attr(*, "names")= chr [1:2] "name" "addr:street"

For example, the names of the restaurants are stored as a string variable called name (second line of output)

The addresses of the restaurants are stored as a string variable called addr:street (second line of output)

Solving problems on your own

• Try to solve a problem for 15 mins
• If you cannot find a solution, have another try in another 15 minutes after taking a break
• Take another break and try again
• Finally, go and aks for help

Resources

• Google
• Stackoverflow
• ChatGBT
• Fellow students
• Me

Collaboration

• This class is an opportunity to learn
• You have two attempts for every assignment
• The final grade for each assignment will be the average of the first and second attempts
• For the second attempt, you should meet up with your assigned team to discuss solutions

Automation and Reproducibility

• Automation is important
• Reduces mistakes
• It is easy to implemet in the long run
• If your data is reproducible, other people can see what you have done: other people can replicate your analyses.

Documentation

• All your should include comments

#Data Cleaning
clean_countries<-subset(life_expectancy2, !(Code %in% weird_labels))
clean_countries_urbanization<-subset(urbanization2, !(Code %in% weird_labels))

Documentation

• All your should include comments

#Data Cleaning
clean_countries<-subset(life_expectancy2, !(Code %in% weird_labels))
clean_countries_urbanization<-subset(urbanization2, !(Code %in% weird_labels))

#Data Cleaning
clean_countries<-subset(life_expectancy2, !(Code %in% weird_labels))
clean_countries_urbanization<-subset(urbanization2, !(Code %in% weird_labels))

#Left Join
new_data<-left_join(clean_countries, clean_countries_urbanization, by = c("Code"="Code"))

Documentation

• All your should include comments

#Data Cleaning
clean_countries<-subset(life_expectancy2, !(Code %in% weird_labels))
clean_countries_urbanization<-subset(urbanization2, !(Code %in% weird_labels))

#Data Cleaning
clean_countries<-subset(life_expectancy2, !(Code %in% weird_labels))
clean_countries_urbanization<-subset(urbanization2, !(Code %in% weird_labels))

#Left Join
new_data<-left_join(clean_countries, clean_countries_urbanization, by = c("Code"="Code"))

• Comments are the what
• Code is the how